Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “transformer based forecasting”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

A Neural Optimizer With Decision-Focused Learning for Optimal Energy Storage Operation

Here, this article introduces a neural optimizer-based framework for optimizing battery energy storage system (BESS) control for grid services, including demand charge and energy cost reduction. By leveraging decision-focused learning (DFL), the proposed framework ensures seamless integration and adaptation, significantly enhancing control performance. A patch time-series transformer is employed for peak load forecasting, incorporating aleatoric uncertainty quantification to account for forecasting uncertainties within the decision-making process. The framework utilizes a solver-in-the-loop approach to generate optimal BESS actions, which are then used to train the neural optimizer-based agent. By co-optimizing both BESS operational modes and output power within the NN, the system achieves improved performance and robustness. After initial training, the forecasting and control models are jointly fine-tuned to account for forecasting errors, further improving decision precision and efficiency through DFL. Case studies are performed to validate the performance of the framework using multiple real-world datasets, demonstrating superior performance in monthly peak load forecasting compared to state-of-the-art models. In addition, the results are compared against existing decision-making approaches. The results demonstrate a reduction in monthly peak forecasting error by approximately 15% across various performance measures and achieve an optimization gap for BESS operation that is about three times smaller compared to existing methods.

Kim, Hyeonjin [Pacific Northwest National Laborato↗
DOE OSTI2024

Short-Term Probabilistic Solar Forecasting via Reinforcement Learning over ECMWF

In this paper, we present an innovative reinforcement learning approach for short-term solar forecasting, leveraging data from the European Centre for Medium-Range Weather Forecasts (ECMWF). The methodology begins with the application of the System Advisor Model (SAM) to transform various ECMWF numerical weather prediction members into predictive photovoltaic power generation. To enhance the precision of deterministic forecasting, we introduce a dynamic model selection algorithm based on Q-learning. This algorithm dynamically identifies and utilizes the most accurate ensemble member for forecasting purposes. Furthermore, we employ a support vector regression surrogate model with a Gaussian distribution to generate probabilistic forecasts, providing a holistic view of solar energy generation uncertainty. To expedite the training process and make it more practical for real-world applications, we integrate a rolling update workflow. This innovative workflow reduces the training period from months to a mere 19 days, making our method highly efficient. Numerical results of the case study show that in comparison to benchmark models, the proposed method improves the deterministic and probabilistic solar forecasting accuracy by up to 40.84% and 48.42%, respectively.

ensemble forecasting↗
DOE OSTI2026

Autoregressive long-horizon prediction of plasma edge dynamics *

Accurate modeling of scrape-off layer (SOL) and divertor-edge dynamics is vital for designing plasma-facing components in fusion devices. High-fidelity edge fluid/neutral codes such as SOLPS-ITER capture SOL physics with high accuracy, but their computational cost limits broad parameter scans and long transient studies. We present transformer-based, autoregressive surrogates for efficient prediction of 2D, time-dependent plasma edge state fields. Trained on SOLPS-ITER spatiotemporal data for the KSTAR tokamak, the surrogates forecast electron temperature, electron density, and radiated power over extended horizons. We evaluate model variants trained with increasing autoregressive horizons (1–100 steps) on short- and long-horizon prediction tasks. Longer-horizon training systematically improves rollout stability and mitigates error accumulation, enabling stable predictions over hundreds to thousands of steps and reproducing key dynamical features such as the motion of high-radiation regions. Measured end-to-end wall-clock times show the surrogate is orders of magnitude faster than SOLPS-ITER, enabling rapid parameter exploration. Prediction accuracy degrades when the surrogate enters physical regimes not represented in the training dataset, motivating future work on data enrichment and physics-informed constraints. Overall, this approach provides a fast, accurate surrogate for computationally intensive plasma edge simulations, supporting rapid scenario exploration, control-oriented studies, and progress toward real-time applications in fusion devices.

autoregressive deep learning↗
DOE OSTI2025

Sensitivity-based voltage constraints for optimal power flow in low-voltage distribution feeders

The optimal power flow (OPF) problem for distribution systems can include network details down to the low-voltage (LV) points of interconnection of individual customers. This paper addresses the implementation of voltage magnitude constraints, and sets forth a practicable approach for capturing the effects on voltage from the switching behavior of loads (e.g., heat pumps, air conditioners, water heaters, or pool pumps) and from the variability of renewable generation (e.g., rooftop solar). The proposed method adjusts the OPF voltage constraints based on forecasts of load and generation upper and lower bounds, in conjunction with sensitivity factors derived from the power flow equations. An illustrative OPF formulation is also provided, which incorporates transformer models that include core loss. We demonstrate that accurate modeling of these LV network components is critical to avoid voltage violations at customer points of interconnection. Furthermore, the ideas are validated through numerical case studies on a realistic distribution feeder.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2024

EV Forecasting-Based Model Predictive Control for Distribution System Congestion Mitigation

The uncoordinated charging of electric vehicles (EVs) in time and space brings congestion issues to the distribution network. This paper proposes an EV charging demand forecasting-based model predictive control (MPC) method for distribution system congestion management. To effectively forecast the time-series EV station charging demand, a hybrid forecasting model that integrates the long short-term memory network (LSTM) and Transformer is proposed. The Transformer-LSTM model is trained using a one-year real historical charging dataset of EV stations to forecast future charging demand in 15-minute intervals. This informs the MPC for distribution network congestion management and minimization of PV curtailment. Numerical results carried out on the modified IEEE 123-bus distribution system demonstrate that the proposed method can effectively resolve line congestion issues through EV smart charging and PV curtailment while outperforming other benchmarks.

ADVANCED PROPULSION SYSTEMS,SOLAR ENERGY↗
DOE OSTI2024

A Scalable Real-Time Data Assimilation Framework for Predicting Turbulent Atmosphere Dynamics

AI-based foundation models like FourCastNet, GraphCast are revolutionizing weather and climate predictions but are not yet ready for operational use. Their limitation lies in the absence of a data assimilation system to incorporate real-time Earth system observations, crucial for accurately forecasting events like tropical cyclones. To overcome these obstacles, we introduce a generic real-time data assimilation framework and demonstrate its end-to-end performance on the Frontier supercomputer. This framework comprises two primary modules: an ensemble score filter (EnSF), which significantly outperforms the state-of-the-art data assimilation method, and a vision transformer-based surrogate capable of real-time adaptation through the integration of observational data. We demonstrate both the strong and weak scaling of our framework up to 1024 GPUs on the Exascale supercomputer, Frontier. Our results not only illustrate the framework's exceptional scalability on high-performance computing systems, but also demonstrate the importance of supercomputers in real-time data assimilation for weather and climate predictions.

Lu, Dan↗
DOE OSTI2025

Deep learning model for fast, science-based forecasting of fluid migration along faults in geologic carbon storage scenarios

Effective long-term geologic storage depends on robust site selection and credible, science-based forecasting of subsurface behavior to ensure storage integrity. For this work, we develop a deep learning–based reduced-order model (ROM) to quantify potential carbon dioxide (CO₂) and brine migration through geological faults. The ROM combines a Transformer model for binary classification and a Stacked Ensemble for regression, trained on a comprehensive dataset generated from 1400 physics-based reservoir simulations. Key geologic and operational parameters—including fault geometry, reservoir structure, and injection conditions—were systematically varied to capture a wide range of fluid migration scenarios. The ROM accurately predicts the onset of migration, cumulative migration volumes of both CO₂ and brine, and associated migration rates, as compared to an independent set of validation simulations, while significantly reducing computational cost compared to traditional simulation methods. Model performance was evaluated across diverse fault configurations, revealing that shallow reservoir geometry and fault angle are among the most influential factors governing migration behavior. Sensitivity analysis using SHapley Additive exPlanations (SHAP) provided interpretability, revealing distinct patterns in how geological and operational features drive transient versus cumulative migration outcomes. The ROM’s ability to rapidly simulate fault migration scenarios enables efficient sensitivity analyses, scenario evaluations, and decision support for site selection and monitoring design. This approach enhances the safety, scalability, and long-term operational performance of geologic carbon storage (GCS) systems by providing a robust, interpretable tool for predicting subsurface fluid migration and assessing fault-related migration potential.

42 ENGINEERING↗
DOE OSTI2025

On the effectiveness of neural operators at zero-shot weather downscaling

Machine-learning (ML) methods have shown great potential for weather downscaling. These data-driven approaches provide a more efficient alternative for producing high-resolution weather datasets and forecasts compared to physics-based numerical simulations. Neural operators, which learn solution operators for a family of partial differential equations, have shown great success in scientific ML applications involving physics-driven datasets. Neural operators are grid-resolution-invariant and are often evaluated on higher grid resolutions than they are trained on, i.e., zero-shot super-resolution. Given their promising zero-shot super-resolution performance on dynamical systems emulation, we present a critical investigation of their zero-shot weather downscaling capabilities, which is when models are tasked with producing high-resolution outputs using higher upsampling factors than are seen during training. To this end, we create two realistic downscaling experiments with challenging upsampling factors (e.g., 8x and 15x) across data from different simulations: the European Centre for Medium-Range Weather Forecasts Reanalysis version 5 (ERA5) and the Wind Integration National Dataset Toolkit. While neural operator-based downscaling models perform better than interpolation and a simple convolutional baseline, we show the surprising performance of an approach that combines a powerful transformer-based model with parameter-free interpolation at zero-shot weather downscaling. We find that this Swin-Transformer-based approach mostly outperforms models with neural operator layers in terms of average error metrics, whereas an Enhanced Super-Resolution Generative Adversarial Network-based approach is better than most models in terms of capturing the physics of the ground truth data. We suggest their use in future work as strong baselines.

17 WIND ENERGY↗
DOE OSTI2026

Spatiotemporal Learning in Power Modules: Wavelet-Enhanced Forecasting of Thermomechanical Degradation

Detecting internal defects in power electronics packages is critical for their performance and reliability, especially under extreme operating conditions, as these defects can lead to catastrophic failure if not properly addressed. Confocal scanning acoustic microscopy (C-SAM) plays a key role in the nondestructive evaluation of bond layer degradation within a power electronics package by detecting defects such as delamination, voids, and cracks. However, accurately quantifying and predicting these defects from C-SAM images remains a significant challenge due to the low noise-to-signal ratio, which typically arises from both imaging process and bond patterns itself. In this paper, we explore machine learning strategies for processing C-SAM images and providing predictive models of defect growth. We use C-SAM images of sintered copper and sintered silver samples, which are obtained under accelerated thermal experiments, as the representative dataset for our study. We investigate the effect of Fourier transforms and wavelet transforms on these datasets to remove high-frequency noise and address noise across multiple scales with histogram equalization to enhance the contrast and improve the visibility of defects. As a result, defect boundaries can be clearly distinguished, enabling more accurate tracking of their growth over time. We then employ different time-series forecasting algorithms on the denoised images to formulate an image-based lifetime prediction model. Statistical models and deep-learning techniques are trained on images obtained in the early stages of thermal shock, and defect growth in the later stages is predicted. Our work serves as a preliminary attempt to improve the accuracy of lifetime prediction models of power electronics packages, which is critical under extreme operating environments.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2025

Novel Deep Learning Transformer Model for Short to Sub‐Seasonal Streamflow Forecast

Accurate short-to-subseasonal streamflow forecasts are becoming crucial for effective water management in an increasingly variable climate. However, streamflow forecast remains challenging over extended lead times, uncertainty in meteorological inputs, and increased frequency and variability in extreme weather and climate events. We implemented a Future Time Series Transformer (FutureTST) model for streamflow forecasting that separately integrates past meteorological and streamflow data while incorporating future weather conditions. FutureTST achieves a mean Nash-Sutcliffe Efficiency (NSE) of 0.82 to 0.67 for 1- to 30-day streamflow forecasts. Incorporating upstream streamflow information improved forecast accuracy by up to 10%. During real-time forecast, FutureTST maintains higher forecast skills of 9.03 for 1-day and 5.74 for 14-day forecasts. In contrast, calibrated process-based hydrological model forecasts become unreliable beyond a 4-day lead time. Our findings demonstrate the potential of FutureTST as a reliable streamflow forecasting tool that offers a valuable addition to operational flood monitoring systems and climate-resilient decision-making.

Ambika, Anukesh Krishnankutty [Oak Ridge National ↗
DOE OSTI2025

A Data-Agnostic, Continuous Machine Learning Framework for Application in High Energy Physics and Beyond: Phase 1 Final Scientific/Technical Report

This Phase 1 effort has focused on the development of continual learning frameworks for use in machine learning, specifically in the applied context of High Energy Physics (HEP). Machine learning (ML) is a transformative technology by which computers, typically through the use of neural networks, are able to perform tasks with proficiency that rivals or surpasses that of human users. Model Degradation & Catastrophic Forgetting are two undesired phenomena which can occur in ML where the performance of a model degrades when either deployed on novel data streams, or trained on novel data which are sufficiently different than the data the models were initially trained on. A natural example where these sorts of effects can be observed is in the performance of detectors in harsh environments, where the detector signature may change over the lifetime of the detector as it ages and deteriorates — precisely what occurs in the experiments conducted in HEP. Real world HEP data is therefore an excellent test-ground and use-case for Continual Learning paradigms, which are techniques used in ML to counteract these problems. Ensemble learning is one such technique, where multiple smaller models are trained on subsets of the overall data and are ensembled together during inference. The intuition behind this technique is that, although there are shifts in the distributions which govern the incoming data streams, these shifts are not expected to be homogeneous or global. If a sufficient diversity in solutions within the various sub-models has been achieved, then at least one sub-model is expected to retain its performance within the overall ensemble. One further strength of this approach is that the architectures of the various models do not need to be identical, and in fact even different modalities of data can naturally be combined in this way. This work focused on applying ensemble learning techniques to derive results using two main datasets, anomaly detection in HEP data & time-series forecasting in semiconductor manufacturing data. Semiconductor manufacturing involves data with surprising similarity to that of HEP (e.g. wafer maps look very similar to digi-occupancy maps) and Cerium Lab’s prominence within the semiconductor industry makes semiconductor manufacturing a natural opportunity for commercialization of this work. Our efforts have led to two strong results. The first is that we evaluated the proposed ensembling techniques using previously proposed machine learning architectures for use in anomaly detection, namely AutoEncoder based models and their derivatives. We also developed new architectures which have not been evaluated in this context before. In fact, this work marks the first use of Vision Transformers for anomaly detection in HEP. Second, we demonstrated that ensemble learning significantly improves model performance in scenarios prone to degradation, validating its effectiveness across both HEP and semiconductor datasets. These results further support ensemble learning as a powerful strategy for mitigating catastrophic forgetting and maintaining robust performance in evolving data environments.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2025

GenAI4UQ: A software for forward and inverse uncertainty quantification using conditional generative AI

We introduce GenAI4UQ, a software package for forward and inverse uncertainty quantification in model calibration, parameter estimation, and ensemble forecasting. GenAI4UQ leverages a generative AI-based conditional modeling framework to address limitations of traditional inverse modeling techniques, such as Markov Chain Monte Carlo (MCMC) methods. By replacing computationally intensive iterative processes with a direct, learned mapping, GenAI4UQ enables efficient calibration of input parameters and generation of predictions directly from observations. The software supports rapid ensemble forecasting with robust uncertainty quantification while maintaining computational and storage efficiency. Built-in auto-tuning of hyperparameters simplifies model training, ensuring accessibility for users with varying expertise. Its versatile conditional generative framework is applicable across diverse scientific domains. While GenAI4UQ offers significant advantages in flexibility and efficiency, users should interpret its uncertainty estimates with caution in data-sparse scenarios, as the model may overestimate uncertainty—an effect common to all surrogate-based approaches including MCMC with surrogate models. Despite this, GenAI4UQ transforms inverse modeling by providing a fast, reliable, and user-friendly solution. It empowers researchers and practitioners to quickly estimate parameter distributions and generate model predictions for new observations, facilitating efficient decision-making and advancing the state of uncertainty quantification in computational modeling.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Leveraging artificial intelligence and advanced food processing techniques for enhanced food safety, quality, and security: a comprehensive review

Artificial intelligence is emerging as a transformative force in addressing the multifaceted challenges of food safety, food quality, and food security. This review synthesizes advancements in AI-driven technologies, such as machine learning, deep learning, natural language processing, and computer vision, and their applications across the food supply chain, based on a comprehensive analysis of literature published from 1990 to 2024. AI enhances food safety through real-time contamination detection, predictive risk modeling, and compliance monitoring, reducing public health risks. It improves food quality by automating defect detection, optimizing shelf-life predictions, and ensuring consistency in taste, texture, and appearance. Furthermore, AI addresses food security by enabling resource-efficient agriculture, yield forecasting, and supply chain optimization to ensure the availability and accessibility of nutritious food resources. This review also highlights the integration of AI with advanced food processing techniques such as high-pressure processing, ultraviolet treatment, pulsed electric fields, cold plasma, and irradiation, which ensure microbial safety, extend shelf life, and enhance product quality. Additionally, the integration of AI with emerging technologies such as the Internet of Things, blockchain, and AI-powered sensors enables proactive risk management, predictive analytics, and automated quality control. By examining these innovations' potential to enhance transparency, efficiency, and decision-making within food systems, this review identifies current research gaps and proposes strategies to address barriers such as data limitations, model generalizability, and ethical concerns. These insights underscore the critical role of AI in advancing safer, higher-quality, and more secure food systems, guiding future research and fostering sustainable food systems that benefit public health and consumer trust.

AI↗
DOE OSTI2024

Modern deep neural networks for Direct Normal Irradiance forecasting: A classification approach

The escalating energy demand and the adverse environmental impacts of fossil-fuel use necessitate a shift towards cleaner and renewable alternatives. Concentrated Solar Power (CSP) technology emerges as a promising solution, offering a carbon-free alternative for power generation. The efficiency and profitability of CSP depend on the Direct Normal Irradiance (DNI) component of solar radiation; hence, accurate DNI forecasting can help optimize CSP plants’ operations and performance. The unpredictable nature of weather phenomena, particularly cloud cover, introduces uncertainty into DNI projections. Existing DNI forecasting models use meteorological factors, which are both challenging to estimate numerically over short prediction windows and expensive to model through data at a sufficiently high spatial and temporal resolution. This research addresses the challenge by presenting a novel approach that formulates DNI prediction as a multi-class classification problem, departing from conventional regression-based methods. The primary objective of this classification framework is to identify optimal periods aligning with specific operational thresholds for CSP plants, contributing to enhanced dispatch optimization strategies. We model the DNI classification problem using four advanced deep neural networks – rectified linear unit (ReLU) networks, 1D residual networks (ResNets), bidirectional long short-term memory (BiLSTM) networks, and transformers – achieving accuracies up to 93.5% without requiring meteorological parameters.

14 SOLAR ENERGY↗
DOE OSTI2025

Forecasting Battery Electrode Performance via Electrochemical Fluorescence Microscopy and Machine-Learning

Predicting lithium-ion battery performance is hindered by microscale electrode heterogeneities invisible to conventional diagnostics. Here, we combine electrochemical fluorescence microscopy (EFM), which maps electronic connectivity by visualizing an electrofluorophore reaction distribution, with a multitask ElasticNet regression to forecast discharge capacity from spatial heterogeneity. Analyzing 196 images from six pilot-scale LiNi 0.5 Mn 0.3 Co 0.2 O 2 cathodes with varying carbon loadings, we extract 62 descriptors that capture morphology and texture. A compact five-feature model predicts capacity across eight discharge rates, achieving a per-target R 2 of up to 0.63 and an overall R 2 of 0.92, with a mean absolute percentage error of less than 2%. This performance rivals impedance-based approaches while avoiding their reliance on postformation data and incomplete electronic network information. Our facile and rapid, image-driven method may enable electrode quality control upstream of costly cell assembly to offer a transformative tool for data-driven battery research and manufacturing.

battery electrodes↗
DOE OSTI2025

Probabilistic Diffusion Models Advance Extreme Flood Forecasting

Extreme floods pose escalating risks in a changing climate, yet forecasting remains challenging due to peak flow underestimation and high uncertainty. We introduce diffusion-based runoff model (DRUM), a probabilistic deep learning (DL) approach that advances extreme flood forecasting across representative basins in the contiguous United States. DRUM outperforms state-of-the-art benchmarks, enhancing nowcasting skill for the top 1‰ of flows in 72.3% of studied basins. Under operational scenarios, DRUM extends reliable lead times by nearly a full day for 20- and 50-year floods. When evaluated with measured precipitation, an ideal condition, recall improves by 0.3–0.4 and the early warning window extends by 2.3 days for 50-year floods. The enhancement potential varies regionally, with precipitation-driven flood zones in the eastern and northwestern US benefiting most, gaining 3–7 days in lead time. These findings highlight the transformative potential of diffusion models as a cutting-edge generative AI technique for advancing hydrology and broader Earth system sciences.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Distributed Cross-Channel Hierarchical Aggregation for Foundation Models

Vision-based scientific foundation models hold significant promise for advancing scientific discovery and innovation. This potential stems from their ability to aggregate images from diverse sources—such as varying physical groundings or data acquisition systems—and to learn spatio-temporal correlations using transformer architectures. However, tokenizing and aggregating images can be compute-intensive, a challenge not fully addressed by current distributed methods. In this work, we introduce the Distributed Cross-Channel Hierarchical Aggregation (D-CHAG) approach designed for datasets with a large number of channels across image modalities. Our method is compatible with any model-parallel strategy and any type of vision transformer architecture, significantly improving computational efficiency. We evaluated D-CHAG on hyperspectral imaging and weather forecasting tasks. When integrated with tensor parallelism and model sharding, our approach achieved up to a 75% reduction in memory usage and more than doubled sustained throughput on up to 1,024 AMD GPUs on the Frontier Supercomputer.

Tsaris, Aristeidis (aris) [ORNL] (ORCID:0000000277↗
DOE OSTI2025

Adoption of AI in the Utility T&D Sector: Use Cases, Consequence, Assessment and Benefits

Digital transformation and utilization of artificial intelligence (AI) in the electric grid are fundamentally changing the industry’s approach to common problems and enabling a broader paradigm shift in grid planning and operations. The change in approach is circularly both enabling and driving modernization, with load growth and reliable management of data center and AI infrastructure shifting away from planning approaches with relatively predictable behaviors and toward a mix of consumer and industrial choices that surpass human cognitive abilities to process. This movement has potential to condition humans to not understand the system on which the AI depends, while requiring it for development of the necessary infrastructure. Approaches which would address most likely grid conditions and events, such as faults, aging of equipment, and weather, now must also account for large loads which shift not based upon weather or time of day, but the computational load. Quantifying computational load is independent of the traditional grid forecasting variables, where a data center’s aggregate load is determined by user and AI system behavior and decoupled from normal grid planning and operations. AI is both the cause and solution for these challenges, with new grid planning tools integrating massive amounts of decisions into frameworks.

24 POWER TRANSMISSION AND DISTRIBUTION↗